Wellx Design System MCP

Wellx Design System MCP

The Wellx Design System MCP enables AI agents to access the single source of truth for Wellx design, including tokens, component specs, rules, and search, ensuring consistent design implementation across surfaces.

Category
Visit Server

README

Wellx Unified Design System

One hub, four surfaces. Everything in this repo is the single source of truth for how Wellx tools and products look; every other artifact is generated from or synced to it.

Surface Consumer Mechanism
Figma "Foundations" + "Wellx Unified Design System" files designers Figma team library (publish from Figma)
tokens/ → @wellx/design-tokens app builds npm package: CSS variables + Tailwind preset
mcp/ → @wellx/design-system-mcp AI agents MCP server: rules, recipes, tokens, search, validate
docs/ (site) humans documentation site — host via GitHub Pages / Vercel; run locally with any static server (npx serve docs)

Using the MCP (agents)

Testing phase — straight from GitHub, no hosting. Add to any repo's .mcp.json (or claude mcp add):

{
  "mcpServers": {
    "wellx-design-system": {
      "command": "npx",
      "args": ["-y", "github:mbarbarydesign/Wellx-design-mcp"]
    }
  }
}

Every new session installs and runs the latest commit of main — push an update and the next session anyone starts is current. Covers Claude Code and Claude Desktop. (Local dev inside this repo: "command": "node", "args": ["mcp/server.mjs"].)

Production phase — remote connector. Deploy the same tool core behind Streamable HTTP on Vercel, then add the URL once as a custom connector in the Claude workspace settings — that extends coverage to claude.ai web/mobile and Cowork. Repos then swap the npx entry for the URL.

Tools: get_principles · get_rules(topic) · get_component_spec(name) · get_tokens(category, mode) · search(query) · validate(code) · get_changelog.

Each portal repo should also keep the thin pointer skill (skill/SKILL.md) in .claude/skills/wellx-design-system/ so agents know to call the server.

Using the tokens (builds)

// tailwind.config.js
module.exports = { presets: [require('@wellx/design-tokens/wellx-tailwind-preset.cjs')], darkMode: ['class'], content: [...] };
/* app entry */
@import '@wellx/design-tokens/wellx-tokens.css';

Updating the system

  1. Designer edits variables in the Figma Foundations file → publishes the library.
  2. Run the sync (scripts/sync-from-figma.mjs with FIGMA_TOKEN, or a Claude session with the Figma MCP) → PR with the token diff + CHANGELOG entry.
  3. Merge → regenerate tokens/*.css|cjs from wellx-tokens.json, bump versions, publish npm packages, redeploy the MCP + docs.
  4. Portal repos pick up tokens via a dependency bump; agents are current immediately via the MCP.

Rules changes (spec/*.md) follow the same PR flow — the MCP serves them live; no agent-side update needed.

Governance

  • CODEOWNERS on tokens/ and spec/ — design lead approves every change.
  • Renaming/removing a token = major version + migration note in CHANGELOG.
  • New components enter spec/components.md only with a reviewed reference implementation.

Documentation site

docs/ is a zero-build static site: What's new, Color tokens, per-component pages (spec + how to use + code), install guide, and the Design System Creator (enter primary/secondary colors, get one design-system.json — guidelines + all tokens — with generated 25-950 scales). docs/one-pager.html is the printable single-page overview.

Ship an update: edit docs/data.js (WHATS_NEW + VERSION), CHANGELOG.md, and the spec/tokens as needed - one PR updates the MCP, the site, and the changelog together.

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured